search

emmo-repo / EMMO

Elementary Multiperspective Material Ontology (EMMO)
Other
65 stars 16 forks source link

Chemical composition and other "compound physical quantities" #30

Closed ahashibon closed 4 years ago

ahashibon commented 4 years ago

in many cases in physics and materials science we need to relate a quantity to multiple "keys". Having a data structure supported for this in the ontology like in #27 may lead to overuse and to "injection" of EAV data paradigms with lower semantic power than is presented in EMMO. here another solution is proposed that alleviates the need for arbitrary keys in data structures. As an example we consider the chemical composition: "$Al 40 %, O 60 %" of some crystal.

here obviously one needs to specify an element, and a corresponding physical quantity (the mass %). instead of defining a map with arbitrary strings, like "Al" --> value "40 %" we can and "O" --> "60 %", we can define that

chemical composition is a physical quantity that has in addition to number and unit an IUPAC chemical element (or molecule).

then the chemical composition of an alloy can simply have properties of all chemical composition instances, without the need for explicit keys.

ahashibon commented 4 years ago

Another option is of course to add a chemical content that has both an element and a quanitity (the mass fraction for example) and having_.

jesper-friis commented 4 years ago

Yes, the latter is the right way to go. I would rewrite it as:

ChemicalComposition hasSpatialPart some ChemicalContent

ChemicalContent isA PhysicalQuantity
ChemicalContent hasSpatialPart exactly 1 ChemicalComponent
ChemicalContent hasSpatialPart exactly 1 Real                 (inherited from PhysicalQuantity)

The subclasses of ChemicalComponent should be all possible chemical components, both pure elements as well as other components like CO2, SiO2, ... When possible, they should refer to an external standard, like the IUPAC Gold Book.

jesper-friis commented 4 years ago

A draft implementation is pushed to https://github.com/jesper-friis/emmo-units and https://gitlab.cc-asp.fraunhofer.de/mat-info/ontology/emmo-units for review.

ahashibon commented 4 years ago

@jesper-friis is it necessary to have chemcial composition part of the emmo-unit or can it be part of a emmo-chemical-composition or a chemical ontology branch.

jesper-friis commented 4 years ago

Currently ChemicalComposition is defined in emmo-si-units, but it is definitely possible to factor it out. It would make sense, since it is (as far as I know) not defined in the SI system.

jesper-friis commented 4 years ago

This is outside the scope of mid-level EMMO, but very interesting to follow up in domain ontologies.